Early menopause may increase the risk of osteoporosis, cardiovascular disease, and all-cause mortality. For approximately 20% of the female population, age at menopause is unknown because they had a hysterectomy without bilateral oophorectomy before their menstrual periods ended. There is some evidence to suggest that hysterectomized women who retain their ovaries are likely to experience ovarian failure earlier that women who go through natural menopause, however most prior studies had important limitations, including small sample size, lack of comparison group, and imprecise measures of menopausal status. In the proposed 5-year study, we will investigate whether hysterectomized women who retain at least one ovary are more likely to experience ovarian failure than women of similar age who have an intact uterus and ovaries. We will recruit 500 premenopausal women aged 30 to 47 years who are undergoing hysterectomy without bilateral oophorectomy at Duke University Medical Center or Durham Regional Hospital, and 500 non-hysterectomized, premenopausal women frequency matched on age and race from gynecology practices in the Durham, North Carolina area. At baseline, all women will have a blood sample drawn and complete an interview focusing on reproductive, hormonal, and lifestyle characteristics that may be related to ovarian function. Serum samples will be analyzed for follicle stimulating hormone (FSH), luteinizing hormone (LH), and estradiol. Women will be recontacted annually to have another blood sample drawn and update questionnaire information, particularly menopausal symptoms and hormone use. Menopausal status at each follow-up will be determined for all women using an algorithm based on measurements of FSH, estradiol, and LH. The primary goals of the study are: 1) to determine whether the risk of ovarian failure during the approximate 4 years of follow-up is greater for hysterectomized women than non-hysterectomized women, and 2) to evaluate the associations between medical, reproductive, and lifestyle characteristics and early ovarian failure. Hysterectomy is the most common non-obstetrical surgery in the United States, with over 600,000 procedures performed each year. The proposed study will contribute greatly to our understanding of the long-term effects of hysterectomy on ovarian function. It should have considerable public health impact given the high frequency of the procedure and the important health consequences of early ovarian failure.